394 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1937 



seed of these hybrids was utilized by Luther Burbank in his widely 

 pubUcized derivation of corn from teosinte in 18 generations. In 

 both Mexico and Guatemala the Indians appreciate that teosinte 

 hybridizes with com and that the eariy generations are unsuitable 

 as grain. Observation has taught them, however, that repeatedly 

 backcrossing the hybrids with the maize parent will in a short wliile 

 eliminate the characteristics contributed by teosinte. It is com- 

 monly stated that if the hybrid seed is sown for three generations 

 with corn it will then become a sort of corn with small ears — a fact 

 which may account for the Aztecs having used the designation 

 teosinte. 



In several regions of Mexico these hybrids of various generations 

 and of backcrosses are found around and in cornfields in great 

 profusion. 



BIRTHPLACE OF MAIZE 



As it happens, the place of the origin of maize may be located 

 with more certainty than the manner of its origin. The exact region 

 of origin of any cultivated plant is not known, but botanists have 

 formulated two guiding principles which, when judiciously applied, 

 produce reasonable results. It has been noted that, other things 

 being equal, a species is less variable on the periphery of its distribu- 

 tion than near its center. Presumably, in wandering from the site of 

 origin, only a part of the inherent variation is carried along any one 

 path, so that at any point on the outer circle of distribution only a 

 part of the variation is represented. If the whole circumference of 

 distribution be traversed, probably the entire range of variation 

 possible to the species would be encountered, but this range would 

 be found all in one place only at the center of origin for the species. 



Applying this principle to locate the origin of maize would put the 

 beginnings of this cereal in Peru, for in the region of Peru and 

 Bohvia occur the greatest number of maize varieties. As with many 

 broad generalizations, there are exceptions to this one. Other things 

 besides germ plasm in a flux and the random selection of genie com- 

 plexes influence the variation of a species. Thus the very nature of 

 the Peruvian terrain would explain the multiplication of maize vari- 

 eties. Cut up into narrow valleys intercommunicating only at the 

 seashore and extending from sea level to great altitudes, even though 

 a single kind of com started its way up all of the Peruvian valleys, 

 the necessary selection for adaptation to different environments and 

 the lack of ready interchange of seed would result in time in each 

 valley having com differing from that of the others. The rugged 

 Peruvian terrain would be expected to produce the multiplicity of 

 com varieties now found there even though a single kind of com had 

 been introduced. 



